Schedule

Intelligent Posting Schedule

Provide data-driven posting time recommendations and help schedule content for maximum first-hour engagement velocity.

Input

The user may ask:

• "When should I post this?"
• "Build me a weekly posting schedule"
• "What's the best time for [content type] aimed at [audience]?"
• Or provide content ready to schedule

Process

Step 1 — Load Timing Data

Read [../../../reference/timing.md](../../../reference/timing.md) for the complete timing reference.

Step 2 — Gather Context

Determine:

• Content type: Post, thread, poll, announcement, etc.
• Target audience: Developers, general tech, consumers, specific niche
• Author's timezone: For converting recommendations
• Posting history: Any known patterns or constraints
• Urgency: Time-sensitive content vs. evergreen

Step 3 — Generate Recommendation

For a single post:

• Recommend the top 3 posting windows with rationale
• Account for content type × timing matrix
• Consider the day of week
• Note: "Be available to reply for 30-60 min after posting" for every recommendation

For a weekly schedule: Build a 5-day plan:

Monday:    [Content type] at [time] — [rationale]
Tuesday:   [Content type] at [time] — [rationale]
Wednesday: [Content type] at [time] — [rationale] ← peak day
Thursday:  [Content type] at [time] — [rationale]
Friday:    [Content type] at [time] — [rationale] (lighter content)

• 3-5 posts per week for quality-focused accounts
• Space posts minimum 2 hours apart on multi-post days
• Threads on Tuesday-Thursday mornings (highest dwell time)
• Lighter content (polls, questions) on Monday/Friday

For a content calendar: Build a 2-4 week plan mixing:

• 1-2 threads per week (highest engagement format)
• 2-3 single posts per week (insights, tips, observations)
• 1 poll per week (engagement boost)
• Daily reply/engagement time (15-30 min)

Step 4 — MCP Integration

If OpenTweet MCP is available:

• Offer to schedule content directly
• Use opentweet_batch_schedule for weekly plans
• Suggest adding high-performers to the evergreen queue

Step 5 — Output

Present:

1. Recommendation with specific times (converted to user's timezone)
2. Rationale for each slot
3. Engagement reminders (reply windows, availability needed)
4. Content type suggestions for each slot if doing a weekly plan

Key Rules

• Always specify timezone
• Wednesday is the peak day — schedule best content here
• Avoid weekends unless the audience is consumer/hobby
• Never schedule without a reply plan — posting and disappearing wastes the first-hour window
• If user can't be available to reply, schedule for the next available window where they can be present

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

• Author: Epistates
• Source: Epistates/sparX
• License: MIT

Install and usage instructions live in the source repository linked above.